Estudio de simulación para comparar varios estimadores de varianza en el marco de la regresión no paramétrica
A simulation study for the comparison of several variance estimators in the nonparametric regression framework
Additional Files
Abstract (en)
We test several difference-based variance estimators in the nonparametric regression model. These estimators have the main advantage of not depending on the smoothing parameters. Furthermore, they also show low computational demand. We mainly use estimators based on ordinary differences, along with estimators based on Hall’s optimal differences. We set scenarios using some regression functions, some sample sizes, and some error distributions. In particular we bring in the use of the half-normal distribution to test the variance estimators under some asymmetric error distributions. Results seem to support the idea that the Hall’s optimal differences estimators not perform better than the others on all sets of scenarios.
Abstract (es)
En este trabajo se prueban varios estimadores de varianza basados en diferencias en el marco de la regresión no paramétrica. Estos estimadores tiene la principal ventaja de no depender de los parámetros de suavización, además de que son poco exigente en términos computacionales. Se usan principalmente estimadores basados en diferencias ordinarias y basados en las diferencias óptimas de Hall. Se crean escenarios utilizando diferentes funciones de regresión, tamaños de muestra y distribuciones de los errores y se introduce el uso de la distribución semi-normal para probar los estimadores de varianza en casos de distribuciones asimétricas de los errores. Los resultados parecen apoyar la idea de que los estimadores basados en diferencias óptimas de Hall no son mejores en todos los escenarios planteados.
References
Altman, N. S. (1992), ‘An introduction to kernel and nearest-neighbor nonparametric regression’, The American Statistician 46(3), 175–185.
Bowman, A. W. & Azzalini, A. (1997), Applied Smoothing Techniques for Data Analysis. The Kernel Approach with S-plus Illustrations, Oxford University Press.
Brown, L. D. & Levine, M. (2007), ‘Variance estimation in nonparametric regression via the difference sequence method’, Annals of Statistics 35(5), 2219–2232.
Buckley, M. J., Eagleson, G. K. & Silverman, B. W. (1988), ‘The estimation of residual variance in nonparametric regression’, Biometrika 75(2), 189–199.
Carter, C. K. & Eagleson, G. K. (1992), ‘A Comparison of Variance Estimators in Nonparametric Regression’, Journal of the Royal Statistical Society, Series B 54(3), 773–780.
Cleveland, W. S. (1979), ‘Robust Locally Weighted Regression and Smoothing Scatterplots’, Journal of the American Statistical Association 74(368), 829-836.
Dette, H., Munk, A. & Wagner, T. (1998), ‘Estimating the Variance in Nonparametric Regression. What is a Reasonable Choice?’, Journal of the Royal Statistical Society, Series, B 60(4), 751–764.
Draper, N. R. & Smith, H. (1966), Applied Regression Analysis, John Wiley & Sons, New York.
Draper, N. R. & Smith, H. (1998), Applied Regression Analysis, 3 edn, John Wiley & Sons, New York.
Eubank, R. L. (1998), Nonparametric Regression and Spline Smoothing, 2 edn, Marcel Dekker, New York.
Gasser, T., Sroka, L. & Jennen-Steinmetz, C. (1986), ‘Residual variance and residual pattern in nonlinear regression’, Biometrika 73(3), 625–633.
Hall, P., Kay, J. W. & Titterington, D. M. (1990), ‘Asymptotically optimal difference-based estimation of variance in nonparametric regression’, Biometrika 77(3), 521–528.
Hall, P. & Marron, J. S. (1990), ‘On variance estimation in nonparametric regression’, Biometrika 77(2), 415–419.
Levine, M. (2006), ‘Bandwidth selection for a class of difference-based variance estimators in the nonparametric regression: A possible approach’, Journal Computational Statistics & Data Analysis 50(12), 3405–3431.
Olaya, J. (2012), Métodos de regresión no paramétrica, Programa Editorial Universidad del Valle, Colombia.
Olmos, N. M., Varela, H., Gómez, H. W. & Bolfarine, H. (2012), ‘An extension of the half-normal distribution’, Statistical Papers 53(4), 875–886.
Rice, J. (1984), ‘Bandwidth choice for nonparametric regression’, Annals of Statistics 12(4), 1215–1230.
Seifert, B. & Gasser, T. (1993), ‘Nonparametric estimation of residual variance revisited’, Biometrika 80(2), 373–383.
Wahba, G. (1978), ‘Improper priors, spline smoothing, and the problem of guarding against model errors in regression’, Journal of the Royal Statistical Society, Series, B 40(3), 364–372.
How to Cite
License
The authors maintain the rights to the articles and therefore they are free to share, copy, distribute, execute and publicly communicate the work under the following conditions:
Recognize the credits of the work in the manner specified by the author or licensor (but not in a way that suggests that, you have their support or that they support your use of their work).
Comunicaciones en Estadística is licensed under Creative Commons Atribución-NoComercial-CompartirIgual 4.0 Internacional (CC BY-NC-SA 4.0)
Universidad Santo Tomás preserves the patrimonial rights (copyright) of the published works, and favors and allows the reuse of them under the aforementioned license.
